Dynamically Updatable Ternary Segmented Aging Bloom Filter for OpenFlow-Compliant Low-Power Packet Processing

OpenFlow, the main protocol for software-defined networking, requires large-sized rule tables and frequent updating. For fast packet processing, rule tables are often implemented with ternary content-addressable memory (TCAM) in the OpenFlow. To deal with TCAM power problems, many network applications employ bloom filters (BFs) to reduce the redundant operations of table-lookup and for low power consumption. However, applying traditional BFs to an OpenFlow switch leads to problems, such as unsupported dynamic update, large space overhead, and the rule-set expansion of ternary data. In this paper, we propose a dynamically updatable ternary segmented aging bloom filter (TSA-BF). The TSA-BF consists of two parts: a segmented aging BF algorithm (SA-BF) and a ternary prefix-tagging encoder (TPE). First, in the SA-BF, we develop an automatic update scheme using the mechanisms of content-aging and buffer-segmenting. The SA-BF ages and deletes its contents automatically, thus eliminating the costly communication overhead and enabling dynamic updating. It also achieves space efficiency by the developed partial-deletion mechanism. Second, in the TPE, we encode ternary prefix-rules into uniquely decodable binary code words. The TPE prevents the rule-set expansion of ternary-data in the OpenFlow environment. Simulation results show that the SA-BF alone can save 37% of space overhead, compared with state-of-the-art techniques. In an environment with the ternary prefix-rules, the TSA-BF can save another 93% of space overhead, compared with the best-performance scheme. Hence, the proposed TSA-BF is highly suited to the requirements of emerging high-performance TCAM-based packet processing in the OpenFlow, which considers dynamic update and power efficiency.